Reducing idling at red lights based on probabilistic prediction of traffic signal timings

Abstract

A predictive optimal velocity planning algorithm is proposed in this paper that uses traffic Signal Phase And Timing (SPAT) information to increase a vehicle's energy efficiency. Encouraged by positive results based on full SPAT information in [1], [2], this current paper focuses on benefits attainable with partial probabilistic information. Availability of signal phase data is categorized into none, real-time only, and full-future knowledge. Dynamic Programming (DP) with full future knowledge of SPAT provides an energy efficiency maximum. The case with no knowledge of phase or timing represents an uninformed driver, and provides an energy efficiency minimum. In between, a signal phase prediction model which could use historically-averaged timing data and real-time phase data is evaluated, as it represents a technology which is available today. Results from a multi-signal simulation indicate that energy efficiency can be increased with probabilistic timing data and real-time phase data.